AI-Vis

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.

====== Seminar: Ethical issues, law & novel applications of AI (MIE 630)======

On Tuesdays, 2pm-3:30pm, Room Gilles Kahn or Sophie Germain, Alan Turing Buidling.

  *  October 2 - **Toward Responsible and Safe AI, Nozha Boujemaa, Dr Inria, director of the DataIA institute (Data Sciences, Intelligence & Society) ** (room Sophie Germain).

**Abstract**: AI technologies are essential for data-driven innovation and digital transformation in many socio-economic sectors. The scientific progress and deployment of AI are increasing sharply. However, several scientific and social challenges remain. In this presentation, I will develop more specifically the challenges we are facing in developing responsible and human values centric AI.

**Bio**: Nozha Boujemaa is a Senior Research Scientist at Inria (the French National Institute for computer science and applied mathematics), Director of DATAIA Institute (Data Sciences, Artificial Intelligence & Society). She is Member of the Board of Directors of Big Data Value Association, Vice-Chair of the AI High Level Expert Group of the European Commission and member of the OECD AI expert group. Nozha is International Advisor for Japanese Science and Technology Agency Program “Advanced Core Technologies for Big Data Integration”, Senior Scientific Advisor for “The AI Initiative“ (Harvard Kennedy School) and President of the Scientific Council of the Institute of Technological Research "SystemX".  Nozha is Knight of the National Order of Merit. Her domains of expertise are Machine Learning, Unsupervised & Semi-supervised Learning, Large-scale Multimedia Retrieval, Content Search, Personalization, Pattern and Object Recognition, Computer Vision and Image Analysis, Transparency and Accountability of data and algorithms covering several application domains. Nozha Boujemaa is the co-author of over 150 international publications and has supervised over 25 PhD students. Previously, Advisor to the Chairman and CEO of Inria in Data Science with concern to the socio-economic impact, Director of Inria Saclay Research Center for 5 years (2010-2015) and Scientific Head of IMEDIA Research Group (Large Scale Multimedia Content Search) for more than 10 years.

  * October 16 - **From artificial intelligence to computational ethics, Jean-Gabriel Ganascia, Sorbonne University, Lip6, Chairman of the COMETS (CNRS Ethical Committee)** (room Sophie Germain)

**Abstract**: With the development of artificial intelligence, it is now possible to design agents that are said autonomous in the sense that their behavior results from a chain of physical causalities from sensors acquisition of signals to action without any human contribution. There are many possible applications of such agents, for instance in transportation, with autonomous cars, or in war, with autonomous weapons. Since human is not present in the loop, many fears the robots animated by such agents be predatory. In order to prevent unsafe behaviors, references to humane values have to be included in the agent programming. More technically, it means that engineers are now designing the equivalent of an "ethical controller" to restrict the robot actions according to moral criteria. To do so, it is necessary to encode different ethical systems, which gives birth to what is called “Computational Ethics”. At the light of the recent autonomous car accident that happened March 2018 in Arizona, we shall detail de different ethical dimensions that such a controller has to satisfy and the technical difficulties that artificial intelligence researchers who deal with computational ethics are facing.

**Bio**: Jean-Gabriel Ganascia is Professor of Computer Science at Sorbonne University, senior member of the Institut Universitaire de France, EurAI – European Association for Artificial Intelligence – fellow and member of the LIP6 (Laboratory of Computer Science of the Paris 6) where he heads the ACASA team. In addition, he chairs the COMETS that is the Ethical Committee of the CNRS and he is member of the CERNA, i.e. the Ethical committee of the Digital Sciences of ALLISTENE, which is the coordination of the French research institutes in computing.

  * October 23 - **Facial recognition: from early methods to deep learning, Stéphane Gentric, Research unit manager, IDEMIA** (room Gilles Kahn)

**Abstract**: 2D Face Recognition is one of the oldest computer vision application. As techniques improve, more complex databases arise, always leaving space for algorithm improvements. This lecture will review the whole face recognition pipeline, how early methods address the main issues and how Deep Learning handles them now. We will present major operational deployments and the most recent performances. Finally, we will discuss current limitations and future research avenues.

**Bio**: Stéphane Gentric is Research Unit Manager at Idemia (ex-Morpho) and a Deep Learning lecturer at ESIEA and Telecom ParisTech. He received his PhD on Pattern Recognition at UPMC in 1999. As principal researcher then team leader, he worked on Fingerprint recognition algorithms, then Face, then Iris and now also Video Analytics. As Senior Expert, he was involved in most of Idemia’s projects in biometrics for the past 15 years, such as the Changi border crossing System as well as NIST benchmarks, or the UIDAI project. His current research interests center around pattern recognition for improvement of biometric systems.


  * November 13 - **Google's AI principles, Ludovic Peran, Google Paris** (room Sophie Germain)


  * November 20 - **Unsupervised Learning on Homogeneous Manifolds, Lie groups and Structured Matrices based on Information Geometry and Souriau Lie Group Thermodynamics, Frederic Barbaresco, THALES** (room Sophie Germain)


  * November 27 - **Law and ethics of autonomous robots, Nathalie NEVEJANS, Lecturer in Law, University of Artois (France)** (room Gilles Kahn).

Nathalie Nevejans is a lecturer in private law at the University of Artois (France), authorized to direct research projects and a member of the CNRS Ethics Committee (COMETS).
Author of numerous articles, participating in events for not only the academic world but also industry, she is one of the few specialists in France on the law and ethics of robotics, artificial intelligence and emerging technologies. She is also a member of the Research Centre for Law, Ethics and Procedures (EA n° 2471), as well as the Institute for the Study of Human-Robot Relations (Etude des Relations Hommes-Robots – IERHR). Her book “Treatise of Law and Ethics of Civil Robotics”, LEH editions (1232 pages) was published in 2017.


  * December 4 - **Algorithmic fairness, Nicolas Usunier, Facebook** (room Sophie Germain)


  * January 8 - **AI@Inria (Research in Artificial Intelligence at Inria), Bertrand Braunschweig, Inria, director of Inria Saclay research center** (room Sophie Germain).


  * January 15 - **Use of personal data and ethics: Vision & challenges from the industry, Sarah Lannes, Research engineer, IDEMIA** (room Sophie Germain).


  * January 22 - **From data protection to data empowerment : how can humans keep the upper hand? Geoffrey DELCROIX, Direction des 
technologies et de l’innovation, CNIL** (room Sophie Germain).


  * January 29 - **Fighting blindness with bionic eyes, Vincent Bismuth, Pixium Vision.** (room Sophie Germain).


  * February 12 - **Kernel methods for genetics - Jean-Philippe Vert, Google Paris.** (room Sophie Germain).


  * February 19 - **Is ethics computable? - Milad Doueihi, philosopher specialist of digital sciences, Université Paris-Sorbonne.** (room Sophie Germain).


  * March 5 - **Learning Prosthetics Design : Function, Shape, Style - François Faure, Anatoscope.** (room Sophie Germain).


  * March 12 - **Gender issues in AI, chatbots & robots - Laurence Devillers, LIMSI & Paris-Sorbonne University.** (room Sophie Germain).
AI-Vis

User Tools

Site Tools

Page Tools
